Photolysis of cellulose and related compounds by 253.7 nm light. II. The role of defects in hydrogen production

Cellobiose and myoinositol, both as received and in physically modified form, were photolyzed at 60.0°C under vacuum with 253.7 nm light. The rates of hydrogen production were compared with the H₂ rate obtained from analogous photolyses of hydrocellulose. The similar physical state and optical properties of cellobiose and hydrocellulose allowed a direct comparison of H₂ production rates, from which it is clear that neither the glycosidic linkage, nor the number of primary or secondary hydroxyl groups, is the basis for observed differences in rates. The relative rates of H₂ production from myoinositol and levoglucosan indicate that strain in the pyranose ring has only a slight effect on the photolysis rate, and further, since published carbohydrate structures show that the local skeletal environment about each hydroxyl in a “defectless” unstrained carbohydrate structure is very similar in all the relevant compounds thus far determined, differences in H₂ rates must therefore arise as a result of possible differences in the hydrogen-bonding environment of the hydroxyl. From an extrapolation of optical properties of simple alcohols in hydrogen-bond-free environments, together with a consideration of possible cage effects during photolysis, it is concluded that the precursors of the H₂ are hydrogen-bond-free hydroxyl groups of the carbohydrates. Confirmation of this hypothesis has been obtained from the H₂ rates observed on photolyzing cellobiose and myoinositol after these compounds had been freeze-dried or recrystallized, which treatment results in a concentration of “defective” hydroxyls differing from the original solid, but is without effect on the overall numbers of “ordinary” hydroxyls.